## Re: devil's algorithm

Expand Messages
• *Sigh*, always the assumptions ruin my hour of triumph! :) CO stands for Corner Orientation. COC1 twists UFR clockwise and UBR counterclockwise. COC2 does the
Message 1 of 46 , Jan 5, 2005
*Sigh*, always the assumptions ruin my hour of triumph! :)

CO stands for Corner Orientation. COC1 twists UFR clockwise and UBR
counterclockwise. COC2 does the same, but with UFL instead of UFR. The
idea is the same with COCs 2 to 7. They all twist one corner clockwise
and UBR counterclockwise. By applying COC1 three times (which is
COT1), we can iterate through the three possible orientations for the
first corner. By applying COC2 three times, with COT1 inbetween each
time, we iterate through all possible orientations for the first two
corners. And so on... up to seven. The last corner we can ignore since
its orientation is decided by the others. So, COT7, which I have
called CO, will cycle through all possible orientations of all the
corners.

The idea is similar with the EO algorithms, except that in EOT1, CO is
called. This means that EOT11 (which by now is getting freakishly
long, and I have called simply "O") cycles through all possible
orientations of all the pieces. So now, we only need to cycle through
all permutations, and apply O once for each of them.

The EP algorithms does that with the edges. EP (EPT12) is a 12 cycle
of edges (EPC12) done 12 times with EPT11 done once inbetween each.
EPT11 is an 11-cycle (EPC11) done 11 times with EPT10 done once
inbetween each... and so on. At the lowest level (EPT2), O is invoked.
For each even cycle, two corners are also swapped, but they're the
same two corners all the time. The edge that is left untouched by the
11-cycle is also left untouched by all the others, etc. (This is
important, I reckon.)

Now, EP is then an algorithm which, given a cube with correct corner
permutation, will eventually produce the solved cube. The CP
algorithms work excactly the same as the EP ones. At the lowest CP
level (CPT2), EP is invoked.

So.. there you go. I think it should work.

wrote:
>
> is there a site where i can find them?

Honor and glory is up for grabs for whoever can find God's algorithm
first, mate!
• Nice! It s been a while. Nice to see some entertainment again. ... [Non-text portions of this message have been removed]
Message 46 of 46 , Sep 20, 2011
Nice! It's been a while. Nice to see some entertainment again.

On Tue, Sep 20, 2011 at 2:59 AM, Alien <rubiks99ca@...> wrote:

> **
>
>
> Happy fun cube
>
>
> The cuber
>
> devil's algorithm
> does anyone know the devil's algorithm?... h_howee
> Jan 5, 2005
> 3:38 am
> Re: devil's algorithm
> ... You mean the shortest way to scramble a cube? Or what is it?... Stefan
> Pochmann
> stefan_pochmann
> Jan 5, 2005
> 5:55 pm
> Re: devil's algorithm
> More like the worst way to solve it. Devil's algorithm is an imagined
> algorithm that, by being applied over and over again, sorts through every
> single possible... Eivind Fonn
> betrayedfiber
> Jan 5, 2005
> 7:15 pm
> Re: devil's algorithm
> How about a devil's algorithm that cycles through every LL situation while
> leaving the F2L intact. That would most likely be easier to find and
> possibly... Chris Sz...
> dishwashersa...
> Jan 5, 2005
> 8:50 pm
> Re: devil's algorithm
> If you find it, please don't post it here. Assuming it looks random, it'd
> take a lot of memory: M = 12!*2^12*8!*3^8/12/1260 = 3.43*10^16 That's the
> minimum... Stefan Pochmann
> stefan_pochmann
> Jan 5, 2005
> 9:35 pm
> Re: devil's algorithm
> Oh, before I forget it: 4325 - The number of years to download it with a
> 1MBit/sec internet connection. So yeah, please don't fucking post it here
> ;-) Stefan ... Stefan Pochmann
> stefan_pochmann
> Jan 5, 2005
> 9:46 pm
> Re: devil's algorithm
> What if I am smart instead? http://www.stud.ntnu.no/~eivindfo/devalg.txtIt's 3,847,762,288,469,010,006,992 moves long, but only 3 KiB large. I bet a
> 1MBit/sec... Eivind Fonn
> betrayedfiber
> Jan 5, 2005
> 10:58 pm
> Re: devil's algorithm
> Yaeh yaeh... ok... remember I was assuming it's random ;-) And hey, where's
> your proof? Ok, not necessary to prove it perfectly formally, but could
> you... Stefan Pochmann
> stefan_pochmann
> Jan 6, 2005
> 12:09 am
> Re: devil's algorithm
> how about god's algorithms? is there a site where i can find them?...
> h_howee
> Jan 6, 2005
> 12:12 am
> Re: devil's algorithm
> *Sigh*, always the assumptions ruin my hour of triumph! :) CO stands for
> Corner Orientation. COC1 twists UFR clockwise and UBR counterclockwise. COC2
> does the... Eivind Fonn
> betrayedfiber
> Jan 6, 2005
> 12:30 am
> Re: devil's algorithm
> Ok, thanks a lot. Very very nice! Actually I've once done something quite
> similar a while ago, so I should've had that idea, too :-) Damn. .. You've
> built it... Stefan Pochmann
> stefan_pochmann
> Jan 6, 2005
> 5:05 am
> Re: devil's algorithm
> ... post. ... Ok, I'm too tired for it tonight, but the point is that these
> two algorithms, cleverly combined over and over again, will be powerful
> enough to... Stefan Pochmann
> stefan_pochmann
> Jan 6, 2005
> 5:10 am
> Re: devil's algorithm
> Ok here's an attempt: Let AC be the cycling alg and AS the swapping alg. My
> Devil's alg does 3*AC, then AS, then 1*AC, then AS, then 4*AC, then AS, and
> so on.... Stefan Pochmann
> stefan_pochmann
> Jan 6, 2005
> 5:47 am
> Re: devil's algorithm
> I would have thought that the "devil's algorithm" would be the shortest way
> to cycle through all N positions of the cube, i.e. a move sequence N-1 moves
> long... _jaap
> Jan 6, 2005
> 7:40 am
> Re: devil's algorithm
> I wasn't aware that it would have to be the shortest one. That makes the
> problem a bit more delicate. :P Stefan: Nice... your's a bit more elegant
> :). The Pi... Eivind Fonn
> betrayedfiber
> Jan 6, 2005
> 10:29 am
> Re: devil's algorithm
> ... I really like the idea of an explicit algorithm for this; the problem
> is distinct from finding a Hamiltonian circuit, though obviously that would
> solve the... mike_go_uk
> Jan 6, 2005
> 12:33 pm
> Re: devil's algorithm
> Your depth-first algorithm is this long (worst case scenario):
> 11,081,777,224,076,574,027,294,514,344,648,207,681,624 moves So I gather
> mine is _slightly_ more... Eivind Fonn
> betrayedfiber
> Jan 6, 2005
> 12:57 pm
> Re: devil's algorithm
> ... True, indeed. Let's not neglect the practicalities ;) Mike...
> mike_go_uk
> Jan 6, 2005
> 1:20 pm
> Re: devil's algorithm
> Excactly. :) I wrote a little script for those who really have time on
> their hands: http://www.stud.ntnu.no/~eivindfo/devalg.php It serves up the
> moves in... Eivind Fonn
> betrayedfiber
> Jan 6, 2005
> 2:48 pm
> Re: devil's algorithm
> ... hands: Thank you! This should be a wonderful resource for anyone new to
> speedsolving. Please, make sure that it is available for the next 120
> trillion... mike_go_uk
> Jan 6, 2005
> 3:11 pm
> Re: devil's algorithm
> Woow! A good candidate for next weeks's FMC ;-) "What is it good for?
> Absolutely nothing! Say it again ..." -Frankie goes to Hollywood (1984)
> </Per> ... into ... Per Kristen Fredlund
> aspiring_to_...
> Jan 6, 2005
> 5:39 pm
> Re: devil's algorithm
> I love it :-) Though I think in the recursion steps you should do all 18
> possibilities, not only six. It gave me the idea of using this with my two
> algs: Let...
>
> --- In speedsolvingrubikscube@yahoogroups.com, h_howee <no_reply@...>
> wrote:
> >
> >
> > does anyone know the devil's algorithm?
> >
>
>
>

[Non-text portions of this message have been removed]
Your message has been successfully submitted and would be delivered to recipients shortly.